EFFECT OF DEMAND UNCERTAINTY ON THE RELATIVE ECONOMICS OF ELECTRICAL GENERATION TECHNOLOGIES WITH DIFFERING LEAD TIMES.

Projections of the demand for electricity in future time periods are subject to considerable uncertainty. Because the demand for electricity depends on factors such as population and economic growth, the more distant the time period for which the projection is made, the greater is the uncertainty. Thus generation technologies with different lead times face demand forecasts with different levels of uncertainty. This fact affects the relative economics of generation technologies with differing lead times. This paper describes a model in which a stylized electric utility faces the decision between a short and a long lead time technology in an uncertain environment. A dynamic programming algorithm is used to determine the least cost investment decision. It is shown that uncertainty can lead to the choice of some short lead time capacity, even when the deterministic solution includes only long lead time capacity. The extent of this effect depends on the nature of the probability distribution of future demands and the relative fuel and capital costs of the two technologies.